Inorg. Chem. 1990, 29, 686-690
686
with the empty orbital on the metal. Rehybridization of the ligating oxygen from sp3to sp2 p would direct a lone pair toward t h e metal, but this seems inconsistent with the normal C-0-C angle for the coordinated THF.
+
Conclusion The essentially planar configuration of the tricoordinate oxygen in bound H20often seen in coordination complexes is now found in an 18-e organometallic aqua complex (2), where the H,O is a 2-e donor. Planar tricoordinate oxygen is also found for the coordinated THF. I t may be that the H-bonds enforce planarity for water, especially where a long Ir-0 bond is present, as here, but the same cannot be true for THF, which is also planar. Arguments about n-bonding and 2-e vs 4-e donor character for ligands of the R 2 0 type should therefore not be based solely on the distinction between planar and pyramidal geometries. The great lengthening of the Ir-O bonds found in 2.THF due t o the trans influence of the hydride ligands is consistent with the lability of the 0 - d o n o r ligands and the unusual activity of the complex in a number of stoichiometric and catalytic reactions. Experimental Section All manipulations were performed under an atmosphere of purified nitrogen with use of standard Schlenk-tube techniques. 'H NMR spectra were recorded on a Bruker WM 250 spectrometer, and IR spectra, on a Nicolet 5-SX FT-IR spectrometer. T H F and hexane were distilled from Na/Ph2C0 and stored under nitrogen over 4-A molecular sieves [ ( ~ o d ) l r ( P P h , ) ~ ] S bwas F ~ synthesized by a method similar to that for the BF, salt29 and recrystallized before use. Dihydridobis( tetrahydrofuran)bis( triphenylphosphine)iridium( 111) Hexafluoroantimonate (1). A red solution of [(cod)lr(PPh3),]SbF6 (1.0 g) in THF (IO mL) was cooled to 0 OC. Hydrogen was bubbled gently through the solution for 10-1 5 min until the solution turned pale yellow. The volume of the solution was then reduced to ca. 4 mL in vacuo. Hexane (40 mL) was added carefully to form a separate layer on top of the T H F solution. After 24 h in a refrigerator, the resulting white crystalline product was filtered out, washed with hexane (4 X IO mL), and dried in vacuo. Yield: 0.79 g (76%). Anal. Calcd for CuH18F61r02P2Sb: C, 48.10; H, 4.40. Found: C, 47.92; H, 4.29. IR (Nujol): ulr+ 2290 cm-I. 'H NMR (CD2CI2,298 K): 8 7.1-7.4 (c, 30 H, Ph), 3.4 (br s, 8 H, CH,), 1.5 (br s, 8 H, CH2), -28.6 (br s, 2 H, Ir-H), (29) Haines, L. M.; Singelton, E. J . Chem. SOC.,Dalton Trans. 1972, 1891.
Preparation of [IrH,(THF)(H,0)(PPh3),~bF6.THF (2-THF) for Crystallography. [IrH2(THF)2(PPh3)2]SbF6(1) (70 mg) was dissolved in 2 mL of slightly moist THF in a small vial that was then submerged in 30 mL of hexane in a Schlenk tube. After several days in a cold room (4 "C), white crystals of 2.THF were formed. X-ray Crystallography of 2.THF. A suitable crystal (0.50 mm X 0.25 mm X 0.20 mm) of 2.THF was sealed in a thin-walled glass capillary that was mounted on an Enraf-Nonius CAD-4 fully automated diffractometer using graphite-monochromated Mo Ka radiation (A = 0.71069 A) and variable scan speed (2.5-IO0/min). The unit cell (see Table I) was determined and refined from 25 randomly selected reflections. The space group, based on the systematic absences observed in the data, was uniquely assigned as P2,/n,with one molecule of CuH,IrP203SbF6 forming the asymmetric unit. The absorption coefficient was ~ ( M Ka) o = 36.78 cm-' and an empirical absorption correction, based on azimuthal scans of three reflections (-2,-7,-2; -2,-7,-l; -2,-7,-3), was applied. The data were also corrected for Lorentz and polarization effects. The structure was solved by a combination of Patterson and difference Fourier syntheses.30 The position of the iridium atom was obtained from the Patterson synthesis. The remaining non-hydrogen atoms were obtained by iterative use of the WFOURIER option in DIRDIF.,' ,411 hydrogen atoms were calculated and input at their idealized positions. The fullmatrix refinement of the non-hydrogen atoms and input of the hydrogen scattering factors resulted in convergence of the crystallographic reliability factor to an unweighted residual of 0.039 and a weighted residual of 0.045. The (101) reflection, having a A(F)/a(F) value of 10.18 was ignored in the refinement. The final difference Fourier showed two electron density maxima close to the Ir atom at 0.2994,0.1995, 0.7769, and 0.5099, 0.1989, 0.7653 with electron densities of 1.9 and 1.8 e/A3, respectively.
Acknowledgment. W e thank the National Science Foundation for support, Professor J. W. Faller and a referee for crystallographic suggestions, and Dr. Marc Zimmer for discussions. Supplementary Material Available: Tables of further bond distances and angles, torsion angles, calculated H atom positions, and thermal parameters for 2.THF (10 pages); a listing of F, vs F, values for 2.THF (25 pages). Ordering information is given on any current masthead page.
(30) TEXSAN-TEXJZAYStructure Analysis Package; Molecular Structure Corp: Woodlands, TX, 1985. (31) Beurskens, P. T. DIRDIF: Direct Methods for Difference Structure-an automatic procedure for phase extension and refinement of difference structure factors. Technical Report 1984/ 1; Crystallography Laboratory: Toernooiveld, 6525 Ed, Nijmegen, The Netherlands.
Contribution from the Department of Chemistry, University of Rochester, Rochester, New York 14627
Hydride Abstraction. The Reaction of IrX(CO)(dppe) with Cp,TaH3 Paul P. Deutsch, John A. Maguire, William D. Jones,* and Richard Eisenberg* Receiued March 21, I989 The reaction of IrX(CO)(dppe) (X = Br, I; dppe- = 1,2-bis(diphenylphosphino)ethane) with Cp2TaH3(Cp = qJ-cyclopentadienyl) is extremely rapid and leads to clean formation offac-IrH3(CO)(dppe) and Cp,TaX, or Cp,TaXL (L = CO, CzH,, C,H,C= CC3HI) in the presence of added ligand. Trapping and isotope-labeling experiments indicate that the reaction does not proceed through production of free IrH,X(CO!(dppe), [IrH(CO)(dppe)], or [Cp,TaH]. the results are consistent with formation of unstable hydride and halide-bridged binuclear intermediates, in which transfer of all hydride and halide ligands occurs before fragmentation into mononuclear species.
Introduction It has been &own that the reaction of Hz with I r X ( C O ) ( d P F ) (1) ( l a , X = Br; Ib, X = I; dppe = 1,2-bis(diphenylphosphino)ethane) leads to rapid and reversible formation of an isomer of IrH2X(CO)(dppe) (2) as the kinetically favored reaction product (eq I).' With time, 2 disappears and another isomer 0020- 1669/90/ 1329-0686$02.50/0
of IrH,X(CO)(dppe) (3) grows in as the thermodynamically favored product. - Adetailed kinetic study of the conversion of 2a into 3a by Kunin et aL2showed that the rate of conversion was ( 1 ) (a) Johnson, C. E.; Fisher, B. J.; Eisenberg, R.J. Am. Chem. Soc. 1983, 105, 7772. (b) Johnson, C . E.; Eisenberg, R. J. Am. Chem. Soc. 1985, 107, 3148.
0 1990 American Chemical Society
Hydride Abstraction
Inorganic Chemistry, Vol. 29, No. 4, 1990 687 H
H-I
x’
r-
I
i 0
2
(1)
\ 3
slow in the presence of excess H2 but was substantially faster with less than 1 equiv of H2. While a reductive elimination/oxidative addition sequence appeared to be operative under high H2 pressure, a bimolecular pathway proposed to involve hydrogen transfer via a dihydride-bridged binuclear species, 4, was operative in the low H2 pressure regime (eq 2).
The high affinity of the iridium(1) system IrX(CO)(dppe) for formal H2 oxidative addition, as indicated by the formation of 3 via the dihydride-bridged binuclear intermediates described above, led to the proposal that 1 might be capable of abstracting H2 from other transition-metal polyhydrides, L,MH,, yielding IrH2X(CO)(dppe) (3) and reactive, coordinatively unsaturated species, L,MH,-2. Whereas reductive elimination of H2 from metal polyhydrides to give coordinatively unsaturated species frequently involves somewhat extreme thermal or photochemical conditions, we reasoned that H2 loss from L,MH, could be effected under mild conditions by dihydride transfer using 1 with the driving force being the formation of an iridium(II1) dihydride complex containing strong Ir-H bonds. The first transition-metal polyhydride chosen for this study was Cp2TaH3 (Cp = v5cyclopentadienyl) because of its known D2 exchange and activation of aromatic C-H bondsS (eq 3). In this paper, we wish to report on the results of this investigation. +
&--I
-
C&
+
D2
A or h v
5
1
2b
+ OIC-I B ’r
5 -ds
(3)
-
Results The reaction of IrX(CO)(dppe) (1) with Cp2TaH3 (5) in benzene-d6 at 25 *C was very rapid and led to the clean formation of IrH3(CO)(dppe) (6)6 and an insoluble precipitate of “Cp2TaX” (7) (eq 4). The reaction occurs rapidly as solutions of 1 and 5
,..‘P> r-P
I b
5
1
+ (2) H
0
Bf
3b
Formation of stable hydride-bridged binuclear complexes is well-known? and similar dihydride-bridged binuclear intermediates have been proposed in closely related s t u d i e ~ . ’ ~Of . ~ particular interest is the reaction of ( V ~ - C ~ H ~ ) R ~ ( Pwith P ~ ~la,) ~which H, resulted in the formation of (qS-CSHS)Re(PPh3),H2and the thermodynamic isomer of IrH2Br(CO)(dppe) (3a).4b Experimental evidence showed that reductive elimination of H2 from ( V ~ - C ~ H ~ ) R ~ ( P followed P ~ ~ ) ~ by H , oxidative addition to l a forming 2a or 3a was nor occurring, and indicated that 3a was formed by direct dihydride transfer from the rhenium atom of ( V ~ - C , H ~ ) R ~ ( P Pto~ the ~ ) ~iridium H ~ center of la via a dihydride-bridged binuclear intermediate, A.
A
(2) Kunin, A. J.; Johnson, C. E.; Maguire, J. A.; Jones, W. D.; Eisenberg, R. J . Am. Chem. SOC.1987, 109, 2963. (3) Venanzi, L. Coord. Chem. Reo. 1982, 43, 251. (4) (a) Drouin, M.; Harrod, J. F. Inorg. Chem. 1983, 22,999. (b) Jones, W. D.; Maguire, J . A. J . Am. Chem. SOC.1985, 107, 4544.
7
6
are mixed, such that the reaction is complete within 3 min when the first NMR spectrum is accumulated. Compounds 1 and 5 react in a 1:l ratio to produce 1 equiv of 6. No evidence for any other Cp2Ta species or IrH,X(CO)(dppe), 2 or 3, was observed by IH NMR spectroscopy. When the reaction was performed at low temperature, similar results were obtained. In toluene-d8 at -60 OC, only 6 and excess 5 were observed in the IH NMR spectrum, and the same species were observed in dichloromethane-d2 at -70 OC. The reaction was rapid even at these temperatures. No intermediates or other species (including unreacted 1) were observed in either of these low-temperature reactions. The reaction of l a with Cp,TaD3 in the presence of Hz was slowed somewhat due to the reversible formation of 2a, as shown in eq 5. The products of this reaction were the “Cp2TaBr” precipitate and 6d3. Analysis of the hydride region of the ‘H NMR spectrum indicated that no incorporation of H2 into the Cp2TaD3reactant or 6d3 product had occurred. The reaction of l a with Cp,TaH3 in the presence of D2 gave similar results, with essentially no deuterium incorporation into the IrH3(CO)(dppe) product. This result is based upon NMR integration of hydrides vs methylene and phenyl ortho protons and upon analysis of the coupling pattern in the iridium-hydride resonance (5) (a) Barefield, E. K.; Parshall, G. W.; Tebbe, F. N . J . Am. Chem. Soc. 1970, 92, 5234. (b) Foust, D. F.; Rogers, R. D.; Rausch, M. D.; Atwood, J . L. J . Am. Chem. SOC.1982, 104, 5646. (6) Fisher, B. J.; Eisenberg, R. Orgunomerallics 1983, 2, 764.
688 Inorganic Chemistry. Vol. 29, No. 4, 1990
Deutsch et al.
111
0
1
2
D
U
6-d,
at 6 -9.48 ppm. It has been shown previously that the coupling in the two outer lines of this hydride resonance, due to J H - H , occurs only in the trihydride complex, 6,6or a symmetry-related dihydride, IrH2(CN)(CO)(dppe), in which both hydrides are trans to the dppe P donors.lb When some deuterium is incorporated into 6, the second-order hydride resonance at -9.48 ppm becomes a simple doublet of doublets. The hydride region of the IH NMR spectrum for the IrH,(CO)(dppe) product of these reactions is shown in Figure 1. As before, no other iridium or tantalum products were observed. In order to confirm the production of "Cp,TaX" (7), IrX(CO)(dppe) and Cp,TaH3 were allowed to react, followed by addition of other ligands to the product mixture. The reaction of added L with 7 was slow, but led to the formation of Cp,TaXL (X = Br, L = C2H4 @a), C3H7C=LC3H7(9a), CO (loa); X = I, L = C3H7C=CC3H7 (9b)). Approximately 0.15 equiv of 8a was formed within 30 min under 1 atm of C2H4, based upon NMR integration of the resonances from 8a vs those from the IrH,jCO)(dppe) product. With L = C3H7C=CC3H7, formation of -0.50 equiv of 9a or 9b was observed after 2 h. The use of CO as added ligand led to recovery of 0.85 equiv of "Cp,TaBr" as 10a after 5 h. When extra ligands, C2H4 or C3H7C=CC3H7, were added before allowing 1 and 5 to react, clean and rapid formation of IrH3(CO)(dppe) and Cp2TaXL (X = Br, L = CzH4, C3H7C= CC,H7; X = I, L = C2H4(Sb),' C3H7C=CC3H7) was observed (eq 6). The amount of Cp,TaXL produced in these reactions
5
1
8 L=C&I,
6
9 L C,H,C=CC,H,
was greater than 0.85 equiv per mole of 5 consumed and of 6 formed. The formation of Cp,TaHL was not observed in any of these reactions. With CO as the added ligand, formation of Cp,TaBr(CO) and a mixture of IrH,(CO)(dppe) and IrH(CO),(dppe) was observed. Independent reaction of IrH3(CO)(dppe) with the adduct IrBr(CO),(dppe) also gave rise to formation of IrH(CO),(dppe) as well as other products, so that the formation of IrH(CO),(dppe) in the reaction between CpzTaH3 and 1 under CO may be due to secondary reaction chemistry and not to a process involving dihydride transfer. To ascertain whether or not monohydride transfer occurs in this system, reactions between Cp,TaHL and IrBr(CO)(dppe) (7) NMR data for Cp,Tal(C,H,) reported by Schrock, R. R.; Sharp, P. R. J . Am. Chem. SOC.1978, 100, 2389.
I
1
-9 0
,
I
,
-10 0 PPY
,
,
,
,
I
1
-I1 0
Figure 1. IH NMR spectra (400 MHz) showing the hydride region of the product IrH,(CO)(dppe) (6) in the reaction of (a) IrBr(CO)(dppe) ( l b ) with Cp,TaH, (5) under N2, (b) l b with 5 - 4 under D2, and (c) I b with 5-d, under H2.
(la) were investigated. A rapid reaction does occur between Cp,TaH(CO) and la in benzene-d,, giving rise to IrH(CO),(dppe) and "Cp,TaBr". However, when mixtures of la with CpzTaH(C2H4)or Cp,TaH(C3H7C=CC,H7) were followed by 'H NMR spectroscopy for 1 h, essentially no reaction was observed. Experiments aimed at determining if initial formation of free IrH,X(CO)(dppe) (3) followed by rapid hydride and halide transfer was occurring were also performed. Little or no reaction of 3a was found to occur with Cp,TaH(CO), Cp2TaH(C2H4),or Cp2TaH(C3H7C=CC3H7) when followed by 'H NMR spectroscopy for 1 h. Discussion Since loss of Hzfrom Cp,TaH, only occurs above 80 OC, the rapidity of the reaction between IrX(CO)(dppe) (1) and Cp,TaH, (5)indicates that a bimolecular process is occurring. It appears that the coordinatively unsaturated complex 1 attacks 5, abstracting all three hydride ligands and transferring the halide ligand to the tantalum center, forming IrH3(CO)(dppe), (6) and Cp,TaX (7). The clean production of 6 is easily detected by IH NMR spectroscopy, and while 7 is not observed in benzene solution, its presence is confirmed by observation of Cp,TaXL in the trapping experiments. As a mononuclear species, 7 is coordinatively unsaturated and would undoubtly seek a donor, accounting for the formation of Cp2TaXL. In the absence of added L, 7 may exist as a halide-bridged dimer, which would create saturated tantalum centers.* The formation of unobservable paramagnetic species such as Cp,TaXz' or Cp2TaXzCpzTaXgacan be excluded on the basis of the sharpness of the 'H NMR spectra. The disproportionation reaction of Cp,TaX giving rise to [(Cp,Ta),] and Cp,TaXz' could be envisioned but can be ruled out because (a) a large amount of Cp,TaXL is formed, >80%, (b) (CpzTa), is not observed in the NMR,9b and (c) CpzTaX2' would not be expected to react with L to give Cp,TaXL. (8) A reviewer has suggested the use of IrMe(CO)(dppe) in these reactions. We have considered this possibility, and are currently investigating the synthesis and chemistry of IrX(CO)(dppe) complexes where X is an
organic group. (9) (a) Antiolo, A.; Fajardo, M.; Otero, A.; Royo, P. J . Organomet. Chem. 1983,246,269. (b) Tebbe, F. N.; Parshall, G.W. J . Am. Chem. SOC. 1971, 93, 3793.
Inorganic Chemistry, Vol. 29, NO. 4, 1990 689
Hydride Abstraction The reaction of 1 with 5 is very fast, and no intermediates are observed at low temperatures (-60 to -70 "C). The only tantalum products observed in any of the reactions are Cp2TaXL. Products such as Cp2TaHL are never found, suggesting that free [Cp2TaH] is not an intermediate in the reaction. Similarly, the iridium product observed in most of these reactions is IrH,(CO)(dppe). An exception to this trend is the formation of some IrH(C0)2(dppe) in reactions performed under CO, but this product has been independently shown to arise from scrambling between iridium centers. The kinetic dihydrogen addition product IrH,X(CO)(dppe) (2) is not a reaction product. This complex is never observed except when the reactions were performed under H2 or D2, where it is formed by reaction of 1 with free H2 according to eq 5 , and not by reaction with 5. Further evidence in support of the fact that 2 is not a reaction product or intermediate is provided by the isotope labeling. The reversibility of H, or D2 addition to 1 forming 2 would also have led to isotope scrambling in these experiments if 2 were one of the initial reaction products. The thermodynamically favored H2 addition product IrH,X(CO)(dppe) (3) is also never observed as a final reaction product or as an intermediate. One could envision the reaction of 1 and 5 proceeding through initial formation of 3 and Cp2TaHL, 8c, 9c, or lOc, followed by rapid subsequent reaction between these two species. However, such a possibility is unlikely on the basis of the results of independent reaction of 3 with %c,9c, and lOc, in which it was found that these compounds reacted only slowly if at all. It is also important to note that free [IrH(CO)(dppe)] is not formed in any of the above reactions. If it were, the isotope-labeling experiments, which were performed under H2and D2,would have led to significant H / D scrambling in the IrH,(CO)(dppe) product, formed by irreversible D2 addition to [IrH(CO)(dppe)] or H2 addition to [IrD(CO)(dppe)], which is inconsistent with the observed experimental results (Figure 1). Even though no intermediates are observed in these reactions, mechanistic insights can be inferred from the observation (or lack thereof) of various products. On the basis of the experimental evidence, it is believed that the attack of 1 on 5 leads to initial formation of a binuclear species, 11, containing two bridging hydrides. It is clear that 11 does not simply break apart by scission of the two Ta-H bonds, since formation of IrH,X(CO)(dppe), 2 or 3, and [Cp2TaH]or Cp,TaHL does not occur. The binuclear species 11 must therefore undergo rearrangement to form a species such as 12 and 13 (eq 7) before finally fragmenting to give
11
12
13
IrH3(CO)(dppe) and Cp2TaX. The rapidity of the reaction and lack of observation of intermediates makes it difficult to determine if the formation of 11 from 1 and 5 and the subsequent conversion of 11 to 12 and 13 are reversible. While these experiments suggest that the reaction proceeds through these binuclear species, these species are unobserved and unstable with respect to the final products. The formation of strong Ir-H and Ta-X bonds provides the underlying force that drives eq 4, especially if the unsaturated Cp2TaX can fill its coordination sphere by dimerizing or binding an added ligand.
Other possibilities for the first step of the reaction between 1 and 5 can be envisioned. Oxidative addition of a tantalum-hydride bond across the iridium center is one possibility. A mechanism such as this has recently been proposed by Casey in the reaction of C P R ~ ( C O ) ~with H ~IrCI(CO)(PPh,),.'O Because of the steric crowding that would occur and the lack of reactivity of 1 with Cp2TaH(C,H,) (812) and Cp,TaH(C3H7C=CC3H7) (9c). it is believed that this oxidative addition mechanism is not operative in our system. The lack of reactivity of 1 with 8c and 9c also indicates that the first step does not involve transfer of a single H+, H', or H- to the iridium center. Concluding Remarks In this study, we have shown that IrX(CO)(dppe) can abstract hydride ligands from Cp,TaH, via polyhydride transfer. Subsequent hydride- and halide-transfer reactions occur in this system, leading to formation of IrH3(CO)(dppe) and "Cp2TaX". Isotope-labeling and trapping experiments indicate that free species such as IrH2X(CO)(dppe), [IrH(CO)(dppe)], and [Cp2TaH]are not produced. This implies that all three hydrides and the halide are transferred between the same tantalum and iridium centers by way of dihydride- and hydride-halide-bridged binuclear intermediates. While dihydride transfer has been observed, the subsequent transfer reactions have prevented observation of chemistry arising from a coordinatively unsaturated species such as [Cp2TaH]. We are currently investigating reactions with other L,MH, systems in hope of observing C-H activation or other chemistry derived from L,MH,-2 species formed by dihydride transfer. Experimental Section All procedures were performed in a nitrogen-filled glovebox or under nitrogen or other appropriate gas on a Schlenk-type vacuum line. All solvents were reagent grade or better and were dried and degassed before use by accepted techniques." Some reagents were used as received, including lithium aluminum hydride (Aldrich), hydrogen (Air Products, 99.9%), deuterium (Air Products, 99.99%), carbon monoxide (Air Products, C P 99.3%). and ethylene (Matheson, C P 99.5%). Boromoform (Aldrich, Gold Label 99+%), bromobenzene (Aldrich, Gold Label 99%), and iodobenzene (Aldrich) were dried and degassed prior to use.ll 4Octyne (Aldrich) was vacuum distilled from sodium prior to use.l' 'H and 31P('HJNMR spectra were recorded on a Bruker WH-400 spectrometer at 400.13 and 161.98 MHz, respectively. Chemical shifts for 'H NMR spectra are reported in ppm downfield from tetramethylsilane but were measured relative to residual IH resonances in the deuterated solvents C6D5H (6 7.150 ppm), C6D5CD2H(6 2.100 ppm), and CDHCI2 (6 5.320 ppm). Chemical shifts for 31P NMR spectra are reported in ppm downfield from external 85% H3P04. A Bruker BVT-1000 temperature control unit was used in the regulation of low temperature when needed and was calibrated against a standard methanol sample. Benzene-d6 (MSD, 99.6% D) and toluene-d8 (Aldrich, 99+% D) were vacuum distilled from sodium benzophenone ketyl, and dichloromethane-d2 (Aldrich, 99.6% D) was vacuum distilled from phosphorus pentoxide. Infrared spectra were recorded on a Mattson Sirius 100 FT-IR instrument. Electron-impact mass spectral analyses were performed on a VG 7035 GC/MS instrument at 20 eV. The complexes IrBr(CO)(dppe)," IrI(CO)(dppe),IL IrH2Br(CO)(dppe) (3e),I2and Cp2TaH(C2H4)I3were prepared according to literature procedures. Preparationof ChTeH3 (5). A modification of the original procedure reported by Green14 was used for the synthesis of this complex. TaCI, (7.7 g, 21 mmol) was slowly added to an ice-cooled mixture of NaCp (7.9 g, 90 mmol) and NaBH4 (2.5 g, 66 mmol) in 100 mL of THF. The resulting mixture was refluxed under N 2 for 4 h and then cooled to room temperature. After careful addition of 1.2 mL of H 2 0 (66 mmol), the mixture was again refluxed for 5 min, followed by removal of the solvents in vacuo. The Cp2TaH3 product was collected from the residue by vacuum sublimation at 110 "C (0.1 Torr). Yield: 2.2 g, 33% isolated (10) Casey, C. P.; Rutter, E. W.; Haller, K. J. J . Am. Chem. Soc. 1987, 109, 6886. ( I 1) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory Chemicals, 2nd ed.; Pergammon Press: New York, 1980. (12) Fisher, B. J.; Eisenberg, R. Inorg. Chem. 1984, 23, 3216. (13) Eichner, M. E.; Alt, H. G.; Rausch, M. D. J . Organomer. Chem. 1984, 264, 309. (14) Green M. L. H.; McCleverty, J. A.; Pratt, L.; Wilkinson, G. J . Chem. SOC.1961,4854.
690 Inorganic Chemistry, Vol. 29, No. 4, 1990
Deutsch et ai.
yield based upon TaCIs. The 'H NMR spectrum of 5, prepared in this way, was identical with that originally reported by Green.I4 lH NMR (C6D6): 6 4.76 (S, IO H, Cp), -1.62 (t. J = IO HZ. 1 H, TaH), -3.02 ppm (d, J = IO Hz, 2 H, TaH). Preparation of Cp2TaD3( 5 4 , ) . The deuterated complex was prepared from 5 - 4 by H/D exchange. A solution of Cp2TaH, (0.38 g, 1.2 mmol), dissolved in C6D6 ( 1 5 mL), was placed under 3 atm of D2 in a FischerPorter bottle and heated to 110 OC for 3 days. Concentration of the solution and addition of hexanes gave 5 - 4 (isolated yield 0.25 g, 65% based upon Cp,TaH,). ' H NMR spectroscopy indicated the product, 5-d,, was pure and was 97% exchanged in Ta-H resonances. Preparation of Cp,TaBr(C,H,) (Sa). CHBr, ( 1 equiv) was added to a stirred solution of Cp,TaH(C,H,) (8c) (10 mg, 0.03 mmol) in benzene (5 mL). After 15 min, the solution was filtered and then taken to dryness, leaving a yellow residue. This residue was taken up into C6D6 for 'H NMR and mass spectroscopic analysis. IH NMR (C&,): 6 4.736
of this solution was removed, the solvent stripped, and C6D6added. 'H NMR spectroscopy revealed the solution to contain both 9c and 5 in a 2:1 ratio. Prolonged heating increased this ratio (3:l at 32 h), but also resulted in significant decomposition. Preparation of Cp,TaBr(CO) (loa). Cp,TaH(CO) (0.17 g, 0.5 mmol) was dissolved in IO mL of toluene. After CHBr, (48 pL, I . 1 equiv) was added, the solution was allowed to stir overnight. The resulting brown solution was filtered, concentrated, and cooled to give a brown solid, which was recrystallized from toluene/hexanes and dried in vacuo (isolated yield 0.10 g, 48%). Anal. Calcd for CIIHIoBrOTa:C, 31.53; H, 2.41; Br, 19.07. Found: C, 30.05; H, 2.67; Br, 21.81.18 IH NMR (C6D6): 6 4.644 ppm (s). IR: vco 1887 cm-I (KBr).,O Mass spectral r Cp,TaslBr, data: no parent ion, but m / e 390 and 392 for c ~ , T a ' ~ Band respectively.I6 Preparation of Cp,TaH(CO) (1Oc). A solution of Cp2TaH3(0.08 g, 0.25 mmol) in toluene (15 mL) was heated to 120 OC under 60 psi of (s,10H,Cp),1.770(t,J=llHz,2H,CH2),1.447ppm(t,J=llHz, C O for 18 h in a Fischer-Porter bottle. After cooling and venting, the 2 H, CH2).IS Mass spectral data: no parent ion, but m / e 390 and 392 solution was filtered under nitrogen and the solvent removed under for c ~ , T a ' ~ B and r CpzTaslBr,respectiveiy.l6 vacuum, leaving Cp2TaH(CO) as a dark purple solid (0.55 g, isolated Preparation of C p z T a I ( C 3 H 7 m C , H 7 (9b). ) The procedure reported yield 65%). 'H NMR (C6D6): 6 4.462 (S, 10 H, Cp), -6.777 ppm (S, 1 H, TaH).22 IR (C6D6): YCO = 1890 cm-l, ~ T ~ -=H 1747 Cm-l. for the methylcyclopentadienyl analogue was f~llowed.'~A solution of NMR Reaction of IrX(CO)(dppe) (1) with Cp2TaH3(5). The general Cp2TaH, (0.16 g, 0.51 mmol), C6HJ (61 pL, 1.1 equiv), and 4-octyne (170 pL, 2.2 equiv) in dioxane (5 mL) was refluxed for I/, h under procedure for the reaction of 1 and 5 will be described. Stock solutions O f l a (3.3 mM in C6D6), l b (3.3 mM in C6D6), and 5 (10 mM in C6D6) nitrogen. The solvent was stripped off, and the solid was extracted with hexanes. Concentrating and cooling the hexane extract gave 9b as a were prepared and stored in a glovebox. yellow solid, which appeared pure by IH NMR spectroscopy (0.14 g, In the glovebox, some of the stock solution of 1 was placed in a vial. isolated yield 50%). Anal. Calcd for CIsH2,ITa: C, 39.44; H, 4.41; I, The vial was then taken out of the glovebox and purged with H2, D,, CO, 23.15. Found: c, 37.03; H, 4.02; 1, 19.22.18 NMR (C6H6): 6 5.067 or C2H4. A color change indicated reversible adduct formation, producing the known complexes IrLX(CO)(dppe) (X = Br, I; L = H,, Dz, (s, IO H,Cp), 3.140 (t, J = Hz, 2 H, a-CH2), 2.433 (t, J = 7 Hz, 2 H, a-CH,), 1.56 (complex multiplet, 4 H, both 8-CH,), 1.018 (t, J = 7 Hz, CO) or the new adducts Ir(C2H4)X(CO)(dppe) (X = Br, Again 3 H, CHJ, 0.924 ppm (t, J = 7 Hz, 3 H, CH,). IR: ucEc = 1772 cm-l in the glovebox 0.05, 0.10, or 0.20 mL of the stock solution of 5 (I/,, I , ( KBr).I9 or 2 @mol,respectively) and enough extra C6D6 to bring the total volume Preparation of Cp2TaBr(C3H7C=CC,H7)(9a). With a procedure up to 0.20 mL was added to an NMR tube, which was then sealed with a rubber septum. Outside of the glovebox the NMR tube was also similar to that used in the synthesis of 9b, Cp2TaH3(0.16 g, 0.51 mmol) flushed with H,, D,, CO, or C2H4,followed by addition of 0.30 mL of was allowed to react with C6HSBr (60 pL, 1.1 equiv) and 4-octyne (240 ,uL, 1.63 mmol) in refluxing dioxane (10 mL) for a period of 4 days. A the gas-saturated solution of 1 (1 pmol) via syringe. The reactions were workup similar to that for 9b gave 9a as a pale yellow powder, yield of then monitored by IH NMR spectroscopy. isolated product 40 mg (16%). This complex was pure on the basis of Low-Temperature Reaction of l a with 5. For these experiments, l a I H NMR spectroscopy. IH NMR (C6D6): 6 5.128 (s, 10 H, Cp), 3.292 and 5 (-2 mg each) were added as solids to an NMR tube fitted with (t, J = 7 Hz, 2 H, (u-CH~),2.449 (t, J = 7 Hz, 2 H, (u-CH~),1.665 a ground-glass joint. On a high-vacuum line, -0.50 mL of C7Ds or (overlapping tq, 2 H, P-CH,), 1.570 (overlapping tq, 2 H, 6-CH,), 1.025 CD2CI2was slowly vacuum distilled into the NMR tube at -196 OC prior (t, J = 7 Hz, 3 H, CH,), 0.941 ppm (t, J = 7 Hz, 3 H, CH,). IR: -U to flame sealing the tube. The sample was subsequently thawed at -78 = 1780 cm-I (KBr).I9 OC in an 2-propanol/dry ice bath and placed in the NMR probe, which was precooled to -60 or -70 OC. Preparation of C p 2 T a H ( C 3 H 7 m C , H 7 )(9c). This was prepared in Reaction of Cp,TaHL (Sc, 9c, of 1Oc) with l a or 3. I n the glovebox, a manner similar to the methylcyclopentadienyl analogue." Cp,TaIsolutions of l a or 3s (-2 mg each) in C6D6were added to NMR tubes (C3H7C=CC,H7) (0.10 g, 0.24 mmol) and excess LiAIH4 (0.10 g, IO equiv) in ether ( I O mL) were refluxed for I h. The ether was removed containing solutions of 8c, 9c, or 1oc in C6D6. The tubes were then sealed under vacuum, and the residue was extracted with toluene or hexanes. with rubber septa and the reactions followed by NMR spectroscopy. The extracts were filtered, concentrated, and cooled to give 9c as an waxy Acknowledgment. We wish to thank the National Science cream-colored solid. 'H NMR spectroscopy indicated the presence of Foundation (Grants CHE 86-03055 and CHE 89-06090) for only one tantalum product, 9c, but revealed the presence of residual support of this work, and the Johnson Matthey Co. for a generous solvents that were not removed by exposure to vacuum. lH NMR loan of iridium trichloride. P.P.D. gratefully acknowledges (C6D6): 6 4.790 (S, IO H, Cp), -0.556 (S, I H, TaH), 2.915 (t, J = 7 Hz, 2 H, a-CH2), 2.786 (t. J = 7, Hz, 2 H, a-CH2), 1.810 (overlapping Sherman Clarke and Elon Huntington Hooker Fellowships. tq, 2 H, P-CH,), 1.741 (overlapping tq, 2 H, P-CH,), 1.04 ppm (unresolved triplets, 6 H, both CH3).I9 The same complex is formed by heating solutions of CpzTaH, with (20) The chloro analogue, Cp,TaCI(CO), has been previously described; see 4-octyne. A toluene solution (25 mL) containing Cp,TaH, (0.16 g, 0.5 ref 21. IH NMR (C6D6): 6 4.75 ppm (s). IR: uco = 1885 cm-I. Characterization of this complex was also reported in ref 9a. mmol) and 4-octyne (1.1 g, IO mmol) was refluxed for 16 h. An aliquot (21) Klazinga, A. H.; Teuben, J. H. J . Orgunomel. Chem. 1979, 165, 31. (22) In ref 9b, Cp,TaH(CO) is prepared by thermolysis of CpzTaH, at 80 OC in benzene under CO. IH NMR: 6 -6.80 ppm (s, Ta-H). IR: uco (1 5) Compare with IH NMR of Cp2TaI(C2H4)and Cp2TaH(C2H4)reported in refs 7 and 13 resoectivelv. CaTaI(C,HA 'H NMR (CAD,): 6 4.80 = 1885 cm-l, u ~ =~ 1750 - cm-'. ~ This complex was also described in (s, lOH,Cp), 1.98'(t,J='11.5'Hz,2H,CH2),1.26ppm-(t~J= 11.5 refs 21 and 5b. From ref 21, 'H NMR (c6D6): 6 4.50 (s, Cp), -6.75 ppm (s, TaH). From ref Sb, IH NMR (C6D6): 6 4.58 (d, J = 1 Hz, Hz, 2 H, CH,). Cp,TaH(C,H,) IH NMR (C&): 6 4.37 (d, J = 0.6 10 Hz, Cp), -6.88 ppm (s, 1 H. TaH). IR (THF): uco = 1890 cm-', Hz, IO H, Cp), -3.50 (m, J = 2.6, 0.6 Hz, 1 H, TaH), 0.63 (td, J = 1 1 , 2.6 Hz, 2 H, CH2), 0.17 ppm (td, J = I I , 2.6 Hz, 2 H, CH,). U T ~ - H = 1745 cm-I. (23) Complexes IrH,X(CO)(dppe), IrD,X(CO)(dppe), and IrX(CO)2(dppe) (16) NO attempt was made to use fast-atom bombardment for ionization. have been described previously: see ref lb. (17) Labinger, J. A.; Schwartz, J.; Townsend, J. M. J . Am. Chem. Soc. 1974, 96, 4009. (24) Addition of -700 Torr of C2H4to 0.4 mL of 10 mM C6D6 solutions of IrX(CO)(dppe) (X = Br, I) causes rapid color change of the solutions (18) Despite numerous attempts at purification of these complexes, better from orange to yellow. For IrBr(C2H4)(CO)(dppe),'H NMR (C6D6): agreement between the calculated values and the found values for the elemental anal ses could not obtained. 6 7.88 (broad s, 4 H, phenyl ortho H), 1.97 (broad s, 4 H, dppe CH2-), (19) Compare with H NMR of ($-CSH4Me),Tal(C,H7C=CC3H7) shown 5.093 ppm (s, excess C2H4).2531P{1H) NMR (C6D6): 6 38.5 ppm (broad s). IR: vco = 1974 cm-l. For IrI(C,H,)(CO)(dppe), 'H NMR (C6D6): in ref 17. 'H NMR (C6D6): 6 -3.4 (t, J = -7 Hz, a-CH,), -2.6 ppm (t, J = -7 HZ, a-CH2). IR: -U = 1775 cm-' (n-C&14). Also 6 7.70 (broads, 4 H, phenyl ortho H), 1.92 (broad d, 4 H, dppe CH2-), compare with the hydride analogue described in ref 17, (9'5.14 ppm ( 5 , excess C Z H ~ ) .''P(lHI ~' NMR (C6D6): 6 39.4 ppm (broad s). IR: uco = 1971 cm-l. Removal of C2H4leads to regeneration of CsH4Me)2TaH(C3H7C=CC3H,). IH NMR (C,D6): 6 -0.2 (TaH), IrX ( C o )(dppe) . 2.9 (t, J = 7 Hz, a-CH2),2.8 ppm (t, J = 7 Hz, a-CH,). IR: uCEc = 1770 cm-'. (25) I n C6D6,uncoordinated C2H4appears at 6 5.24 ppm. I).23324
r
